Global and Planetary Change 172 (2019) 1–14

Contents lists available at ScienceDirect

Global and Planetary Change

journal homepage: www.elsevier.com/locate/gloplacha

Invited review article Arctic kelp forests: Diversity, resilience and future T ⁎ Karen Filbee-Dextera, , Thomas Wernbergb,e, Stein Fredriksenc, Kjell Magnus Norderhaugd, Morten Foldager Pedersene a Norwegian Institute for Water Research (NIVA), Gaustadalléen 21, 0349 Oslo, Norway b UWA Oceans Institute and School of Biological Sciences, University of Western Australia, Crawley 6009, WA, Australia c University of Oslo, Department of Biosciences, PO Box 1066, Blindern, N-0316 Oslo, Norway d Institute of Marine Research, Nye Flødevigveien 20, NO-4817 His, Norway e Department of Science and Environment (DSE), Roskilde University, DK-4000 Roskilde, Denmark.

ARTICLE INFO ABSTRACT

Keywords: The Arctic is one of the most rapidly changing places on Earth and it is a sentinel region for understanding the Seaweed range and magnitude of planetary changes, and their impacts on ecosystems. However, our understanding of Climate change arctic coastal ecosystems remains limited, and the impacts of ongoing and future climate change on them are Polar largely unexplored. Kelp forests are the dominant habitat along many rocky Arctic coastlines, providing struc- Sea ice loss ture and food for economically and ecologically important species. Here we synthesize existing information on Borealization the distribution and diversity of arctic kelp forests and assess how ongoing changes in environmental conditions could impact the extent, productivity, and resilience of these important ecosystems. We identify regions where the range and growth of arctic kelp are likely to undergo rapid short-term increase due to reduced sea ice cover, increased light, and warming. However, we also describe areas where kelps could be negatively impacted by rising freshwater input and coastal erosion due to receding sea ice and melting permafrost. In some regions, arctic kelp forests have undergone sudden regime shifts due to altered ecological interactions or changing en- vironmental conditions. Key knowledge gaps for arctic kelp forests include measures of extent and diversity of kelp communities (especially northern Canada and northeastern Russia), the faunal communities supported by many of these habitats, and the role of arctic kelp forests in structuring nearby pelagic and benthic food webs. Filling in these gaps and strategically prioritizing research in areas of rapid environmental change will enable more effective management of these important habitats, and better predictions of future changes in thecoastal ecosystems they support and the services that they provide.

1. Introduction timing and magnitude of primary production, and driving shifts in the structure and function of marine communities (Grebmeier et al., 2006; The effects of humans are pervasive and are transforming natural Nelson et al., 2014). As a result, the entire Arctic region has been de- ecosystems and biogeochemical cycles on global scales (Halpern et al., signated an ocean warming hotspot (Hobday and Pecl, 2014). Impacts 2008; Waters et al., 2016). There is, however, great regional variation of rapid environmental change on arctic ecosystems has broad sig- in the nature, magnitude, and direction of these changes (Burrows nificance due to both the global uniqueness and large geographic extent et al., 2011; Krumhansl et al., 2016), and it is only by understanding of the region, and because they may act as a sentinel for other eco- these geographical intricacies that we can begin to grasp the full extent systems experiencing slower rates of change (Pecl et al., 2014; Hobday of our footprint on the planet. Currently, the Arctic is warming 2–4 and Pecl, 2014). Despite this, most Arctic coasts remain relatively un- times faster than the global average and is now one of the most rapidly explored, and the extent and resilience of coastal ecosystems are poorly changing regions in the world (IPCC, 2014). Marine ecosystems along understood, as are the ongoing and future impacts of climate change on Arctic coasts are experiencing increases in sea temperatures, dramatic them. Understanding changes to arctic ecosystems is especially critical declines in sea ice, and increased input of freshwater (Wassmann and because borealization (i.e., the northward shift of temperate commu- Reigstad, 2011; Coupel et al., 2015; Acosta Navarro et al., 2016; Ding nities) could squeeze out high arctic ecosystems altogether, resulting in et al., 2017). These changes are altering carbon cycling, affecting the the planetary loss of an entire climate zone (Fossheim et al., 2015;

⁎ Corresponding author. E-mail address: [email protected] (K. Filbee-Dexter). https://doi.org/10.1016/j.gloplacha.2018.09.005 Received 21 March 2018; Received in revised form 28 August 2018; Accepted 6 September 2018 Available online 12 September 2018 0921-8181/ © 2018 Elsevier B.V. All rights reserved. K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

Kortsch et al., 2015). conditions south of these limits (AMAP, 1998). Here we define ‘arctic Kelps are large brown seaweeds that occur on rocky coasts kelps’ as kelps occurring within the boundaries defined by the Arctic throughout the Arctic (Wernberg et al., 2019). Many (or most) kelps are Monitoring and Assessment Program (AMAP). AMAP originally defined important foundation species that create habitat (forests) for numerous Arctic boundaries in 1991 as regions north of the 10 °C July isotherm. fish and invertebrates (Christie et al., 2009; Norderhaug and Christie, These boundaries have since been expanded to include some areas that 2011; Teagle et al., 2017), provide food to marine communities through correspond to political boundaries of member nations of the Arctic high production and export of detritus and dissolved organic material Council (e.g., coastal shelf of Iceland, Norwegian northwest coast, (Krumhansl and Scheibling, 2012; Renaud et al., 2015; Abdullah et al., Hudson Bay, and the Aleutian Islands) (AMAP, 2017). We used this 2017; Filbee-Dexter et al., 2018 in press), and store and sequester definition because monitoring programs, assessments and decision- carbon (Krause-Jensen and Duarte, 2016). Currently, information on making on pollution and climate change in Arctic regions often use the distribution, diversity, stability, and function of kelp forests is AMAP boundaries. However, despite our inclusive definition of the missing for large portions of the Arctic (Wiencke and Clayton, 2009; Arctic, much of this manuscript focuses on kelp forests at higher lati- Krumhansl et al., 2016; Wilce, 2016). tudes within the AMAP region where kelps face the most extreme Arctic A recent global analysis of records of kelp abundance over the past 5 conditions and where globally unique species compositions are found. decades showed that kelp forests are changing in many regions of the world (Krumhansl et al., 2016). At the warmest edges of their range, 2.2. Distribution, growth forms and evolution of arctic kelps sudden shifts from kelp forests to reefs dominated by low-lying turf- forming algae have been increasingly documented over the last decade Although kelps range along most Arctic coasts, sparse records of (Filbee-Dexter and Wernberg, 2018). Along other temperate coasts, kelps in some parts of the Arctic have been attributed to a lack of hard native kelps are being replaced by invasive kelps or other seaweeds substrata (Kjellman, 1883; Wilce, 2016). Only about 35% of the Arctic (Wernberg et al., 2019), or are being heavily overgrazed by sea urchins basin is rocky substrate and shallow coastal areas and inner Arctic (Filbee-Dexter and Scheibling, 2014). In many of these regions, declines fjords are often dominated by sediment due to glacial run off and river in kelp abundance are partly explained by the direct and indirect effects deposition (Leont'yev, 2003; Lantuit et al., 2012), which limits the of warming sea temperatures (Ling et al., 2009; Catton, 2016; Filbee- presence of macroalgae. In areas with suitable substrate, dense kelp Dexter et al., 2016; Wernberg et al., 2016). Considering the widespread forests can extend from the intertidal zone down to depths of 30–40 m changes throughout the temperate and tropical range of kelp and the depending on light conditions, wave regime, and grazing intensity ongoing environmental changes occurring in the Arctic, the fate of (Wernberg et al., 2019). The deepest recorded kelp was observed at arctic kelps in this era of rapid change is a critical gap in our knowledge 60 m depth in Disko Bay, Greenland (Boertmann et al., 2013). In high of arctic marine ecosystems. Arctic regions, available light and sea ice further restrict this depth Here we synthesize existing information on the distribution, bio- range and the upper sublittoral zone is a barren, low salinity environ- mass, and dominant species of arctic kelp forests. We explore some of ment that is constantly impacted by sea ice and meltwater (Wiencke the services provided by arctic kelps and identify missing baseline and Clayton, 2011). measures of their extent. We analyze changes in the sea ice extent and The diversity of kelp in the high Arctic tends to be lower than in temperature conditions for known locations of kelp, and explore how temperate kelp forests (Wiencke and Clayton, 2011). Genetic evidence recent and future changes in these and other conditions could impact indicates that most kelps reinvaded the Arctic from the Atlantic Ocean their growth, reproduction, and survival. Finally, we highlight key gaps ~8000 years ago following the last ice age, which eliminated benthic in our understanding of these ecosystems, and suggest strategies for flora from most current Arctic subtidal regions (Wulff et al., 2011). As a future research. result, most arctic kelps have optimal growth temperatures that exceed those experienced during the Arctic summer and many of these species 2. Hidden blue forests of the arctic therefore also thrive along warmer, temperate coasts (Wiencke and Amsler, 2012). In the high Arctic especially, kelps tend to be morpho- 2.1. Bounds of arctic marine ecosystems logically smaller compared to their southern range limits (e.g., Kuznetsov et al., 1994; Kuznetsov and Shoshina, 2003; but see Borum Arctic and temperate marine ecosystems are separated by a moving et al., 2002). However, kelps still form dense canopies in some regions boundary, generally defined by latitude, sea ice cover, light variability, (e.g., western Alaska and northern Norway) and provide most of the and the locations of the polar front and other ocean currents algal biomass and the largest three-dimensional biogenic structure on (Piepenburg, 2005). The locations of these boundaries can be seasonal, rocky coasts in Arctic regions (Wiencke and Amsler, 2012). In fact, unpredictable, and can shift with climate change. A precise and uni- these lush underwater forests are particularly striking in the Arctic, versally accepted geographical definition of ‘arctic marine ecosystems’ where terrestrial coasts are barren and ice scoured with little three- therefore does not exist, and different southern limits for arctic marine dimensional structure. ecosystems are used in the literature (Zenkevitch, 1963; Piepenburg, The species pool is relatively young, with only one truly arctic en- 2005; Gattuso et al., 2006; Wilce, 2016). For example, so called ‘Arctic demic kelp, Laminaria solidungula (Kjellman, 1883; Zenkevitch, 1963; conditions’ (ice scoured intertidal zones, ocean temperatures < 0 °C, Wilce and Dunton, 2014). All other kelp species found in Arctic regions and months with little to no daylight) extend below the Arctic circle also extend into sub-arctic and northern temperate waters and include along the coasts of Greenland and Eastern Canada, which are influenced Alaria esculenta, Agarum clathratum, Eualaria fistulosa, Laminaria digitata, by the cold southward moving Labrador and Greenland currents, but Laminaria hyperborea, Nereocystis luetkeana, Saccharina latissima, Sac- are restricted to above the Arctic circle along the coasts of northern charina longicruris, Saccharina nigripes, Saccorhiza dermatodea, Alaria Norway, Iceland and in the southern Bering sea, which are influenced elliptica, and Alaria oblonga (the latter 2 are only found in Russia) by the warmer northward moving Gulf Stream and North Pacific cur- (Fig. 1, Table 1). There is currently taxonomic confusion regarding rents, respectively (Wilce, 2016). The convergence of cool waters from some arctic species; S. nigripes, for example, has often been mis- the Arctic Ocean and warm waters from the Atlantic and Pacific Oceans identified as L. digitata, and appears to be restricted to Arctic or sub- occurs around 65°N on the east coast of Greenland, 80°N west of arctic conditions, although more information on its distribution is Svalbard, 76 °C in the Barents Sea, in the Bering Strait, 63°N in the needed (McDevit and Saunders, 2010). In 2006 a new species of kelp eastern Canadian Arctic Archipelago, and then slightly north between Aureophycus aleuticus was collected from Kagamil Island, Aleutian Is- Baffin Island and the west coast of Greenland (AMAP, 1998). However, lands, but its classification within the order Laminariales is still unclear other factors such as sea ice, light, and glacial run-off also create Arctic (Kawai et al., 2013). New DNA barcoding techniques show promise for

2 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

Fig. 1. Photographs of select kelps from high Arctic regions: (a) Laminaria solidungula, (b) Alaria elliptica, (c) Saccharina longicruris, (d) Saccharina nigripes, and (e) Saccorhiza dermatodea (Guiry and Guiry, 2017). clearing up misidentifications caused by diverse growth morphologies covered by snow (Mundy et al., 2007). Subtidal habitats in the Arctic of kelps in arctic conditions (McDevit and Saunders, 2010; Bringloe can therefore be without light for much of the year. Studies from NE et al., 2017). Greenland illustrates this; the annual surface irradiance (PAR) in Young Sound (74° 18′ N) amounts to ca. 6100 mol photons m−2, but the ice- 2.3. Adaptations to arctic conditions free period is limited to August and September so that the amount of available light at 10 and 20 m depth is only 234 and 40 mol photons −2 −1 Kelps in arctic environments are challenged by extremely low water m yr , respectively (Borum et al., 2002). temperatures, periods of low salinity, and extreme variability in light The marked seasonal variation in light availability in the Arctic caused by large annual variations in day length, light intensity, and sea concentrates primary production into a short period and creates strong ice cover. In their northernmost range, kelps live in temperatures at the seasonality in the growth of kelp (Chapman and Lindley, 1980; Dunton point of freezing sea water during polar nights (e.g., NE Greenland, and Jodwalis, 1988; Borum et al., 2002; Makarov et al., 2008). Arctic Borum et al., 2002; Franz Joseph Land, Shoshina et al., 2016). Day- kelps are well adapted to these long periods of darkness or low light length ranges from 24-h sunlight in mid-summer to several months of conditions. Studies on S. latissima and L. solidungula show that these total darkness during winter (Hanelt, 1998). The low angle of the sun species store most of the carbon obtained during the short summer and periods of complete darkness mean that high Arctic areas only period and subsequently use these reserves to form new blades during receive 30–40% of the light received in the tropics on an annual basis. the succeeding period of almost darkness (Chapman and Lindley, 1980; The long period of darkness during winter is further extended in areas Dunton and Jodwalis, 1988; Borum et al., 2002). Remarkably, the peak with partial or complete sea ice cover, especially if the ice is thick or growth period for Alaskan L. solidungula was from February to April

Fig. 2. Kelp locations (red) within AMAP Arctic boundary line (orange). Gray shading shows max- imum sea ice extent, blue shading shows continuous permafrost (90–100% cover), discontinuous perma- frost (50–90%), and sporadic and isolated patches of permafrost (< 50%) (2016 National Snow and Ice Data Centre, https://nsidc.org/data/docs/fgdc/ ggd318_map_circumarctic/). Eroding coasts (yellow) and stable coasts (light green) in regions with sea ice were differentiated according to the Arctic coastal classification scheme developed by Lantuit et al. (2012).

3 .Fle-etre al. et Filbee-Dexter K. Table 1 Species composition, depth limit and biomass (wet weight per m2) of arctic kelp forests. Bolded names indicate dominant species. (−) is not reported.

Location Site Year Depth limit Species Latitude, Longitude Kelp WW (g m−2 ) Mean ± SE Reference sampled (m) (n)

Canada Hudson and Kangirsuk L. solidungula S. longicruris 60.0373, −70.1796 11.8 ± 1.3 (25) (Sharp et al., 2008) Ungava Bay Hudson and Basking I 10 L. solidungula S. longicruris 59.9848, −69.9478 2.9 ± 0.2 (25) (Sharp et al., 2008) Ungava Bay Labrador sea E. Port Markham 2003 30 A. clathratum A. esculenta 52.3667, −55.7333 801.8 (Adey and Hayek 2011) Labrador sea Tilcey I 2003 20 A. clathratum A. esculentaL. digitata S. 52.2167, −55.6333 1808.8 (Adey and Hayek 2011) dermatodea S. latissima Labrador sea South Cove 2003 30 A. clathratum A. esculenta S. dermatodea S. 53.2167, −55.6333 4109.8 (Adey and Hayek 2011) latissima S. longicruris Baffin Bay Walls I, Cape St. Charles 2003 12 A. clathratum A. esculenta L. digitata S. 52.2167, −55.6167 1903.4 (Adey and Hayek 2011) dermatodea S. latissima Hudson and Tuvalik Pt. 12 A. clathratum A. esculenta L. solidungula S. 60.0568, −69.6745 8.4 ± 1.1 (25) (Sharp et al., 2008) Ungava Bay groenlandica S. longicruris Hudson and Pikyuluk I 12 A. esculenta L. digitata L. solidungula S. longicruris 59.9868, −69.9337 9.2 ± 2 (25) (Sharp et al., 2008) Ungava Bay

Greenland Baffin Bay Qaanaaq 2009 A. clathratum S. latissima S. longicruris 77.4667, −69.2500 15.0 ± 2.6a (Krause-Jensen et al., 2012) Baffin Bay Dundas 77.5500, −68.8667 14.9 ± 0.8a (Krause-Jensen et al., 2012) Baffin Bay Uummannaq 2009 33 A. clathratum S. latissima 70.6667, −51.6000 24.1 ± 4.0a (Krause-Jensen et al., 2012) Labrador sea Disko Bay 69.4833, −53.6333 18.8 ± 0.9a (Krause-Jensen et al., 2012) Labrador sea uuk 2008 30 A. clathratum A. esculenta S. longicruris 64.1333, −51.6167 18.0 ± 1.1a (Krause-Jensen et al., 2012) Labrador sea Eqip Sermia 2009 27 A. clathratum S. latissima 69.7500, −50.3500 12.6 ± 2.8a (Krause-Jensen et al., 2012) 4 Norway Norwegian Sea Finnøy-Håvær V 2012 20 A. esculenta L. hyperborea S. latissima 62.8203, 6.5472 1141.1 ± 349,1 (Christie et al., 2014) Norwegian Sea Finnøy-Håvær N 2012 A. esculenta L. hyperborea S. latissima 62.8252, 6.5546 1301.0 ± 360,3 (Christie et al., 2014) Norwegian Sea Vega-Ivarsbraken 2012 20 A. esculenta L. hyperborea S. latissima 65.6764, 11.5494 1589.7 ± 377,7 (Christie et al., 2014) Norwegian Sea Vega-Bubraken 2012 20 A. esculenta L. hyperborea S. latissima 65.6802, 11.5984 712.7 ± 246,2 (Christie et al., 2014) Norwegian Sea Vega-Igerøy 2012 A. esculenta L. hyperborea S. latissima 65.6901, 12.1310 788.3 ± 133,9 (Christie et al., 2014) Norwegian Sea Senja-Sjursvika 2012 20 A. esculenta L. hyperborea S. latissima 69.0956, 16.7792 818.4 ± 174,5 (Christie et al., 2014) Norwegian Sea Senja-Stongeland 2012 20 A. esculenta L. hyperborea S. latissima 69.0427, 16.8795 307.8 ± 69,0 (Christie et al., 2014) Norwegian Sea Senja-Halvardsøya 2012 20 A. esculenta L. hyperborea S. latissima 69.1599, 16.8958 864.3 ± 115,9 (Christie et al., 2014) Norwegian Sea Senja-Kjerringbergnes 2012 20 A. esculenta L. hyperborea S. latissima 69.3110, 16.8978 741.8 ± 135,9 (Christie et al., 2014) Norwegian Sea Senja-Månesodden 2012 20 A. esculenta L. hyperborea S. latissima 69.3111, 16.8978 1038.7 ± 92,3 (Christie et al., 2014) Norwegian Sea Senja-Lemmingsvær 2012 20 A. esculenta L. hyperborea S. latissima 69.0270, 16.9326 561.2 ± 125,3 (Christie et al., 2014) Norwegian Sea Hekkingen I 2016 10 A. esculenta L. hyperborea S. latissima 69.6167, 17.8860 21,976.0 ± 2967,0 (Filbee-Dexter et al., 2018) Barents Sea Kongsfjorden 2013 20 A. esculenta L. digitata L. solidungula S. 78.9833, 11.9632 4614.0 (Bartsch et al., 2016; Hop dermatodea S. latissima et al. 2016) Barents Sea Finnmark-Kongsfjord 2012 20 A. esculenta L. hyperborea S. latissima 70.6991, 29.4393 691.7 ± 110,7 (Christie et al., 2014) Global andPlanetaryChange172(2019)1– Barents Sea Posangerfjord – – 4.1 ± 1.8 (Christie et al., 2014) Barents Sea Finnmark-Bøkefjord 2012 20 A. esculenta L. hyperborea S. latissima 69.8525, 30.1300 703.5 ± 163,9 (Christie et al., 2014)

Russia Barents Sea Cape Abram 15 S. latissima 69.0210, 33.0226 613.3 (Shoshina et al., 2016) Barents Sea Cape Mishukov 6 A. esculenta S. latissima 69.0595, 33.0429 183.3 (Malavenda and Malavenda, 2012) Barents Sea Belokamenka Bay 6 S. latissima 69.0777, 33.1807 836.7 (Malavenda and Malavenda, 2012) Barents Sea Cape Retinskiy 6 A. clathratum L. digitata S. latissima 69.1122, 33.3793 1550.0 (Malavenda and Malavenda, 2012) White sea Ostrov Asafiy 1973 9 S. latissima 66.4210, 33.6559 1922.0 (Myagkov, 1975) White sea Nikolskaya Bay 8 L. digitata S. latissima 66.2167, 33.8333 5232.0 ± 1201,0 (Plotkin et al., 2005)

(continued on next page) 14 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

under full ice cover (Dunton, 1985), and the production of new lamina in S. latissima from Young Sound (NE Greenland) occurred under ice cover and in complete darkness, likely based on re-allocation of carbon from the old lamina or stipe (Borum et al., 2002). Many kelp species can also cope with multi-year sea ice, which can cause severe mechanical damage to benthic organisms in the intertidal and upper subtidal zone (Krause-Jensen et al., 2012; Dayton, 2013; ( Wilmers et al., 2012 ) ( Dean et al., 2000a ) ( Wilmers et al., 2012 ) ( Konar et al., 2017 ) ( Konar et al., 2017 ) Reference ( Dunton and Schell, 1986 ; Dunton et al., 1982 ) ( Duggins et al., 1989 ) ( Duggins et al., 1989 ) Shoshina et al., 2016). Most kelp forests recover from sea ice damage through high reproduction and recolonization of the scoured substrate. Keats et al. (1985) found, for example, that populations of A. esculenta recovered within a few years after having been removed by ice-scour in the uppermost reaches of its range. However, Konar (2013) found slow ) Mean ± SE b b

−2 recolonization in clearing experiments on kelps in the Boulder Patch (< 10% recolonization after 7 years), which is much slower than rates in many temperate kelp forests. 1908 ± 372 SE 900 ± 200 SE 12,645 ± 4999 12,645 ± 4999 (n) 262.0 2920 ± 1810 2628 ± 1912 Kelp WW (g m 3. Known locations of arctic kelps

Data on the current extent and distribution of kelps in the Arctic is not available. To overview the observational data record of kelps in subarctic and Arctic seas we compiled records of kelps over the last 2 centuries, within the AMAP boundaries, from primary literature, mu- seum collections, dive logs, Arctic expeditions, coastal monitoring, and local ecological knowledge from Inuit and northern communities (N = 1179 records, Fig. 3). The spatial extent of these ecosystems ranged from 100 s of km2 of kelp forests to small patches of kelps within inner fjords and boulder patches along sedimentary coasts. The number of kelp records decreased with latitude, with the northernmost ob- 51.5521,-178.4067; 51.6102,-177.0966; 51.8619,-175.1848; 52.3509,-170.8579 60.3327, −147.7644 52.0563,177.4398 51.5961, −178.6748 70.3208, −147.5833 52.7790,-169.3972; 53.2908,-167.9203 3523 ± 674 SE 51.6102,-177.0966 51.5043,178.4812 Latitude, Longitude servations of kelp forests > 80° N at Svalbard, Norway and Franz Jo- seph Land, Russia (Shoshina et al., 1997; Bartsch et al., 2016). Most records were from northern Norway, western Greenland, eastern Ca- nada, and northwestern USA. The earliest records of arctic kelps were S. latissima spp. spp. from the Canadian high Arctic during expeditions in search of the spp. spp.

spp. Northwest passage (Lee, 1980). Other early records come from Kjellman (1883), who published the first comprehensive review of S. latissima Laminaria Laminaria polar benthic algae based on expeditions from Sweden via Norway to Laminaria spp. spp. Novaya Zemlya, and into the Siberian sea, Russia, and Rosenvinge (1893, 1899), who described the algal flora in Greenland a decade later. Dive research on arctic kelp forests was first conducted in Greenland, E. fistulosa E. fistulosa E. fistulosa , Laminaria L. solidungula Laminaria Laminaria E. fistulosaLaminaria E. fistulosa Canada and USA by Wilce (1963), Chapman and Lindley (1980), and Dunton et al. (1982). It is worth noting that these historical records represent a baseline and may not reflect current kelp distributions. A. clathratum Ondonthalia setacea Ptilota serrata Laminaria longipes A. cribosum A. cribosum, A. cribosum A. esculenta A. clathratum Ondonthalia setacea Ptilota serrata Laminaria longipes E. fistulosa E. fistulosa Species Extreme variation in environmental conditions occur within the AMAP arctic boundaries. Large regional differences in coastal condi- tions are strongly driven by the cover of sea ice and the presence of permafrost (frozen soil, rock, or sediment) (Lantuit et al., 2012). To

Depth limit (m) capture this variability in our description of arctic kelps, we grouped information from our observational data into 3 general categories: (1) kelps on stable coasts with sea ice, (2) kelps on unstable, eroding coasts with sea ice, and (3) kelps on coasts with little to no sea ice. 2016 – 1998 – – 1980 7 1987 30 20161987 – 30 Year sampled 3.1. Kelps on stable arctic coasts with sea ice

Stable, rock bound coasts and fjord systems in Arctic areas with seasonal cover of sea ice can support luxurious kelp forests, although their vertical distribution is limited by ice scour (shallow) and light (deep). These areas are expected to experience pronounced changes in E. fistulosa . environmental conditions when sea ice retreats. Although this should increase overall primary productivity along these coasts, the species Chuginadak I Amchitka I Unalaska I Site composition of algae currently found in these Arctic regions may be lost permanently if more temperate-adapted algal communities push northward and outcompete kelps that are adapted to seasonal sea ice (Krause-Jensen and Duarte, 2014). ( continued ) In the northern Barents Sea, kelp forests of mixed A. esculenta, L. SE of dominant species Dry weight.

a b digitata and S. latissima occur within high latitude fjords off Svalbard, Aleutian Islands Tanaga I, Adak I, Atka I, Gulf of Alaska Knight Island Aleutian Islands Ogliuga I Aleutian Islands Kiska I USA Beaufort sea Boulder patch Aleutian Islands Umnak I/Anangula I, Aleutian Islands Adak I Location

Table 1 the western White Sea, and Franz Joseph Land (Kuznetsov et al., 1994;

5 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

Fig. 3. Photographs show examples of arctic kelp forests: (A) Laminaria solidungula in the Beaufort Sea, Alaska, USA (Ken Dunton), (B and C) Laminaria hyperborea in Malangen fjord, Norway (Thomas Wernberg, Karen Filbee-Dexter), (D) Eularia fistulosa Aleutian Islands, Alaska (Pike Spector), (E) Saccharina latissima under sea ice in Kangiqsujuaq, Canada (PBS, 2017), (F) Laminaria digitata in Svalbard, Norway (Max Schwanitz), (G) Saccharina latissima, S. longicruris, Alaria esculenta, Laminaria solidungula in northern Baffin Island, Canada (Frithjof Küpper), and (H) Laminaria hyperborea in Murmansk, Russia (Dalnie Zelentsy).

Cooper et al., 1998; Bartsch et al., 2016; Fig. 3F,H). Luxuriant stands of 3.2. Kelps on eroding, permafrost bound arctic coasts with sea ice L. digitata, L. solidungula, S. dermatodea, and A. clathratum were ob- served within fjords in western Novaya Zemlya (Shoshina and Scattered low relief, rocky coasts in the eastern Siberian, Laptev, Anisimova, 2013). In the northernmost regions around Svalbard and Beaufort, and Chukchi seas, and the Canadian high Arctic have tem- Novaya Zemlya, the arctic endemic kelp L. solidungula is found in inner peratures and light conditions that should support kelp (Krumhansl and fjords and areas that receive cold polar currents (Svendsen, 1959; Hop Scheibling, 2012), but observations are rare in these regions et al., 2012; Shoshina and Anisimova, 2013). (Zenkevitch, 1963; Lee, 1973; Wilce and Dunton, 2014; Wilce, 2016). The west coast of Greenland is largely rockbound and dominated by These coasts are more permanently icebound compared to other Arctic sub-littoral kelp forests from Cape Farewell in the south (59° N) to regions– especially in the Beaufort, eastern Siberian, and Laptev seas – Smiths Sound in the north (> 80° N, Rosenvinge, 1893, 1899). The and the seafloor is often covered in sediment due to intense glacial run western Greenland kelp forests are dominated by S. longicruris north of off. Low salinity, high levels of sedimentation, and sparse substrate 62° N and by S. latissima south of this latitude, while other species such make kelps and other macroalgae poorly developed (Taylor, 1954; as L. solidungula, A. esculenta, Agarum clathratum, S. nigripes and S. Leont'yev, 2003; Dayton, 2013). As a result, kelps along these coasts dermatodea are present, but less conspicuous (Rosenvinge, 1899; face harsh conditions such as extensive sea ice scour, long periods of Krause-Jensen et al., 2012). The kelp forests in western Greenland are darkness, variable salinity, turbidity, and/or low temperatures (Wilce, narrow and shallow in the north, but become broader, more abundant, 2016). The associated macroalgal communities in many of these regions and extend deeper in the south due to less ice cover (Krause-Jensen have distinct species compositions compared to other regions of the et al., 2012). In some parts of Greenland, high densities of sea urchins Arctic, possibly because they are less connected to nearby temperate or a lack of hard bottom restricts the extent of the kelp forests (Krause- communities due to outflow of polar currents from the north to south Jensen et al., 2012). The kelp populations in eastern Greenland tend to along their coasts (Wilce and Dunton, 2014). In the Alaskan Beaufort be situated deeper, have less biomass per unit area and grow more Sea, kelps are found in scattered rocky habitats in shallow waters slowly than those on the west coast (Borum et al., 2002; Krause-Jensen (5–10 m depth) along the mainly sedimentary coast. Research on kelps et al., 2012), which may be due to lower water temperatures, longer in this area are from the ‘Boulder Patch’ (71° N), where L. solidungula periods with ice-cover, and more heavy scour by pack ice. S. latissima forms beds intermixed with A. esculenta and S. latissima on shallow and A. esculenta appear to be the dominant species along most of the cobbles and boulders (Wilce and Dunton, 2014; Fig. 3A). These isolated east coast (recorded as high as Danmarks Havn (75° N)), while L. soli- kelp communities contain about half of the 140 macroalgal species dungula, S. nigripes, S. longicruris and A. clathratum are present, but less found in the Arctic. The Boulder Patch has been studied since 1978 and abundant (Rosenvinge, 1899). revisited in 14 separate years between 1978 and 2012, over which time In Hudson Bay and eastern Canada, sea ice extends below the Arctic the species composition has remained relatively static (Wilce and circle due to the influence of the cold Labrador current. S. latissima, A. Dunton, 2014). clathratum, A. esculenta, and L. solidungula have been documented be- In the northwestern high Canadian Arctic, low availability of rocky tween Ellesmere Island and Labrador, and along coasts in Lancaster substrate and a harsher climate support smaller, fragmented kelp for- Sound, Ungava Bay, Hudson Bay, Baffin Bay, and Resolute Bay ests (Lee, 1980). This region of the Canadian Arctic commonly supports (Table 1). These ecosystems can be highly productive in some areas, L. solidungula, which has been observed as high as 74.5° N. with luxuriant beds of 15-m long S. latissima observed in Frobisher Bay, Along sedimentary coasts in northeastern Russia, observations of and beds containing a biomass of 19 kg wet weight m−2 of A. esculenta kelps are limited to a handful of records, namely, S. latissima off measured in Ungava Bay (Sharp et al., 2008). Kelp forests have also Amderma, mainland Russia, Kotel Nyy Island (Cooper et al., 1998), and been documented in eastern Chukchi Sea from Norton Sound to north of along the Russian coast of Chukchi Sea (Zenkevitch, 1963); L. soli- the Bering Strait along the west coast of Alaska (70 and 71° N; Phillips dungula on islands in the Laptev Sea and within bays in the Siberian Sea and Reiss, 1985). (Cooper et al., 1998), and S. latissima, L. solidungula, S. nigripes, A. el- liptica and A. oblonga in the Kara sea (Zenkevitch, 1963; Guiry and Guiry, 2017).

6 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

3.3. Kelps in arctic regions with little to no sea ice Boulder Patch, USA (Hamilton and Brenda, 2007; Włodarska- Kowalczuk et al., 2009; Wilce and Dunton, 2014). Kelp canopies can Kelp forests in the Norwegian Sea, the Barents Sea, and the northern create favourable conditions for some understory species and were Pacific (Aleutian Islands and northern Gulf of Alaska) have high upper shown to provide predation refuge for juvenile cod in Newfoundland, limits of biomass compared to other arctic kelp forests (Table 1; Canada (Gotceitas et al., 1995) and rockfish and ronquils in the Gulf of Fig. 3B,C,D). These regions have little to no sea ice and ocean tem- Alaska (Dean et al., 2000b). Traditional knowledge from northern peratures that are warmer than other Arctic regions due to the influence communities in Greenland reported higher arctic cod catches in areas of the Gulf Stream or the Pacific Current. Kelp forests in some of these near kelp forests compared to other areas (Krause-Jensen and Duarte, regions (e.g., the Gulf of Alaska) are highly influenced by environ- 2014). Despite these reports, the smaller size and patchy nature of kelps mental conditions on land, namely high freshwater inputs from melting in some Arctic regions may reduce their importance as habitat forming permafrost and melting glaciers that creates strong clines in salinity in species compared to temperate forests. Kelp also has cultural value for coastal areas (Spurkland and Iken, 2011; Lind and Konar, 2017). Kelp in northern peoples and features in their traditions and stories. It is a other regions with little to no sea ice appear to be more influenced by traditional food for Inuit, who harvest it from under sea ice during low biological factors than by environmental conditions. Many kelp forests tide (Wein et al., 1996) and can be used by farmers as fertilizer or to are strongly impacted by herbivorous sea urchin populations, which cattle feed (Reedy and Katherine, 2016). can increase with the loss of higher level predators (e.g., crabs, cod, Kelp-derived organic material constitutes a significant component of otters) (Doroff et al., 2003; Filbee-Dexter and Scheibling, 2014). Im- coastal primary production, often forming the base of benthic food portantly, kelps currently found in areas with little to no sea ice may webs in nearby habitats (Dunton and Schell, 1987; Fredriksen, 2003; represent future scenarios for other Arctic regions. Krumhansl and Scheibling, 2012). Direct consumption rates on most Along the western and northern coast of Norway, and along low- high arctic kelps are unknown, but are likely lower than those along lying, rock-bounded coasts within the Murmansk region of Russia, L. temperate and subarctic coasts, as herbivores tend to be less abundant hyperborea dominates the exposed coasts (Fig. 3B,C, Table 1) and kelp and the digestion of algae hypothesized to be less energy efficient in forests can obtain biomasses up to 21 kg fresh weight m−2 (Fig. S1). In colder ecosystems compared to warmer ecosystems (Floeter et al., the mid-1970s, high densities of the green sea urchin Strongylocentrotus 2005; Konar, 2013; Wilce, 2016). Konar (2007) deployed grazer ex- droebachiensis destructively grazed kelp forests and created extensive clusion cages in experimental clearings in kelp forests in the Beaufort urchin barrens, restricting kelps to exposed regions or shallow surf Sea, Alaska, and found that the overall increase in algal recruitment due zones (Leinaas and Christie, 1996). Currently, regional recovery of kelp to grazing was < 1% of the total area cleared. Similarly, the sea urchin forests is occurring following decreases in sea urchin populations due to S. droebachiensis, a key grazer of kelps along temperate coasts in the reduced urchin recruitment in the south (Fagerli et al., 2013) and in- North Atlantic (Filbee-Dexter and Scheibling, 2014), is confined to creased crab predation in the north (Fagerli et al., 2015). shallow waters in the south western Barents Sea (Murman coast), lo- In the North Pacific Ocean, surface canopy forming kelps E. fistulosa calized patches in Jan Mayen (Gulliksen et al., 1980), Novaya Zemlya and N. luetkeana and subsurface kelps (A. clathratum, A. esculenta, (Nordenskiøld, 1880) and southern parts of Svalbard (Gulliksen and Costaria costada, L. digitata, and S. latissima) form forests along the Sandnes, 1980), and is rare or absent around Franz Josefs Land and the Aleutian Island chain, the northern Gulf of Alaska coast and the Laptev and Kara Sea (Levin et al., 1998). Exceptions to this pattern of northeastern coast of Russia. E. fistulosa dominates surface canopies in low grazing pressure at higher latitudes include kelp forests in the the Aleutian Islands and E. fistulosa and N. leutkeana in southeast Alaska Aleutian islands and northern Norway, where high consumption rates that can grow from > 30 m depth. Subsurface kelps tend to be com- by sea urchins have been recorded (Estes and Duggins, 1995; Leinaas petitively dominant in both regions (Duggins, 1980, Dayton 1975). Kelp and Christie, 1996). forests in the northern Gulf of Alaska occur within the largest fresh- Kelp carbon contributions to marine organisms in coastal environ- water discharge system in North America, and experience strong gra- ments can be substantial. On average, around 80% of the kelp pro- dients of salinity due to substantial glacial inputs. The amount of glacial duction globally (91% for the Boulder Patch in the Beaufort Sea) enters melt is increasing with climate change, further lowering salinity and coastal food webs as detritus, through detachment or exudation of negatively effecting kelps in these areas (Lind and Konar, 2017). In dissolved organic carbon, which is exported to adjacent ecosystems on contrast, kelp forests along the shores of the Aleutian Islands are more beaches and deeper offshore areas (Krumhansl and Scheibling, 2012). influenced by biotic interactions. These coasts have alternated between Macroalgal-derived carbon can be used by benthic herbivores and kelp forests and urchin barrens for over a century (Estes et al., 2004). predators, while upper trophic level fishes and marine mammals gen- Shifts between these two ecosystem states are driven by changing erally use phytoplankton-derived carbon (McMeans et al., 2013). Stable abundances of sea otters, which are major predators of the sea urchin isotope analyses show kelp carbon contributed 57% to nearshore fish Strongylocentrotus polyacanthus (Estes and Duggins, 1995). Evidence populations in the Gulf of Alaska (von Biela et al., 2016), 15–75% to from the region suggests that kelp forests established in 1911 after sea rock greenling, predatory sea stars, and cormorants in the Aleutian Is- otter populations rebounded (Estes et al., 1978). The recovered kelp lands (Duggins et al., 1989), 0–42% for diverse marine predators in forests (E. fistulos and Laminaria spp.) were maintained for decades, Baffin Island, Canada (McMeans et al., 2013), and 50% to mysid crus- until otter populations declined again due to predation by killer whales taceans in the Beaufort Sea (Dunton and Schell, 1987). The latter pre- (Doroff et al., 2003; Estes et al., 2004), once again limiting kelp forests datory snails are a critical food source for higher trophic levels such as to exposed areas and shallow depths, which act as refuges from grazing fish, whales, and birds, indicating the high importance of kelpasa (Konar and Estes, 2003). primary producer (Dunton and Schell, 1987). A comprehensive understanding of the nature and extent of kelp 4. Ecosystem services provided by arctic kelp forests subsidy to other arctic benthic, pelagic, and terrestrial ecosystems is still lacking, and the magnitude and importance of kelp exported from Kelps can provide extensive substrate for colonizing organisms, and shallow coasts to deeper habitats is a debated topic of on-going research their canopies create habitat for a number of marine plants, fish, and (Renaud et al., 2015). In the subarctic and Arctic regions, most research invertebrates (Teagle et al., 2017). The flora in arctic kelp forests canbe has focused on the vertical influx of phytoplankton- or zooplankton- diverse and has been described in detail for some high Arctic regions derived organic matter as the main source of carbon in benthic systems. (e.g., Wilce and Dunton, 2014; Küpper et al., 2016). Diverse fish, in- In Greenland, the primary production of kelps and other benthic algae vertebrate and epiphytic communities are found in kelp forests in can contribute to > 20% of the total primary production in shallow Svalbard, Norway, the Aleutian Islands, the Gulf of Alaska, and the coastal areas. However, at depths > 15 m this production was largely

7 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

Fig. 4. (a) Global trends in predicted increase in mean summer (July 21 to Sept 21) surface temperature from 2016 to 2036 according to IPCC models. Kelp locations are shown in red within AMAP Arctic boundary line (blue). (b) Rate (y−1) of historic and (c) rate of projected warming of peak summer temperature (Aug to Sept) calculated on basis on linear trend analysis for all for all 1° latitude radius buffers around each kelp forest record. insignificant compared to that of phytoplankton and benthic micro- 5.1. Temperature algae (Krause-Jensen et al., 2007). The magnitude of, and timing by which, kelp-derived carbon enters arctic ecosystems is especially in- Temperatures in the Arctic are projected to increase by 3–4 °C by the teresting because climate change is triggering earlier phytoplankton end of the 21st Century under realistic warming scenarios (IPCC, 2014; blooms in the Arctic, creating temporal mismatch between pelagic Huang et al., 2017). Currently, kelps in Arctic waters experience low primary production and some higher trophic level species that syn- temperatures with little seasonal variation. Water temperatures rarely chronize their life cycle or behaviour to this pulsed source of energy exceed 5 °C in summer in the high Arctic, but may reach 10 °C during (van Leeuwe et al., 2018). In light of this mismatch, understanding summer in the southern-most parts of Arctic or where warm ocean other sources of arctic primary production during food-limited periods currents affect local climate. Average temperatures may be below 0°C is becoming critical. with a variation as small as ± 1 °C in high latitude places affected by Knowing the residence time of kelp detritus in Arctic environments cold currents (e.g., Igloolik, Northwest Territories, Canada (Bolton and is important in light of increased interest in blue carbon sequestration Lüning, 1982); Young Sound, eastern Greenland (Borum et al., 2002); worldwide (Krause-Jensen and Duarte, 2016). In the Canadian high Franz Joseph Land, Russia (Shoshina et al., 2016)). Arctic, large amounts of macroalgal detritus have been observed on the To explore prior and ongoing temperature changes in the vicinity of seafloor in sheltered fjords (Küpper et al., 2016). In northern Norway documented locations of arctic kelp, we related these to maps of surface (70°N), pulses of whole kelp blades rapidly reached deep-fjord com- temperature for the region. We calculated average temperature mea- munities (> 400 m depth) during the spring shedding of old L. hy- sures from 1986 and 2016 at each of our kelp locations using historical perborea lamina (Filbee-Dexter et al., 2018). If kelp material degrades IPCC temperature maps (IPCC, 2014, accessed through slower and remains intact longer in colder arctic environments, it may gisclimatechange.ucar.edu). Around each kelp location we averaged be more likely to be buried and sequestered in ocean sediments than the mean summer (July to September) temperature over this 20-year kelp carbon produced at lower latitudes. period within a buffer radius of 1° latitude, which corresponded tothe spatial error associated with locations of early records. We also calcu- lated the magnitude and rate of the predicted increase in mean summer 5. Kelps in a sentinal region of change temperature at each location using climate model forecasts for 2016 to 2036 (IPCC, 2014). We used the model based on the conservative Key changes that will influence kelps in the Arctic include elevated greenhouse gas emission scenario B1, which predicted a conservative temperatures (Najafi et al., 2015; Wang et al., 2017), decreased cover increase of 1.1–2.9 °C by 2090–2099 relative to 1980–1999 (SRES, and thickness of sea ice (Arctic Monitoring and Assessment 2000). Programme., 2011; Parkinson and Comiso, 2013; Ding et al., 2017), The mean summer temperature across all kelp locations has in- reduced salinity, and increased turbidity (IPCC, 2014; Günther et al., creased by 0.35 °C ( ± 0.20) per decade over the period from 1986 to 2015). Other environmental changes that could impact kelps are altered 2016 (Fig. 4a) and is predicted to increase by 1.09 °C ( ± 0.59) per nutrients levels and increased UV radiation. Reduced sea ice and decade over the next century (Fig. 4b). Predicted temperature increases warming could also bring in invasive species by increasing shipping are least pronounced for kelps along the coasts of Greenland and traffic or warm water species migration (Miller and Ruiz, 2014), which eastern Siberia, and most pronounced in the Barents Sea, Beaufort Sea, could impact kelp communities. The cumulative impact of these stres- and Canadian high Arctic, suggesting that changes to kelp forests due to sors will likely affect kelp growth rates and periods severely, but ulti- warming will first occur in these regions. mately depends on their nature and strength, the interactions between Based on temperature tolerance and growth optima of most arctic them, and the ways in which different kelp species acclimate and/or kelp species, warmer temperatures should increase growth rates adapt to new conditions (Harley et al., 2012). (Müller et al., 2009; Shoshina et al., 2016). The optimum growth

8 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14 temperature for most arctic and cold-temperate kelp species range from weekly from 2006 to 2016 (http://nsidc.org/, NOAA, accessed 2017). 10 to 15 °C (Wiencke and Amsler, 2012; Roleda, 2016), and growth at We constrained our measures to this period because years prior to 2006 0–5 °C is typically only 25–30% of growth at their optimum tempera- had lower resolution spatial measures for coastal regions. At each kelp ture (e.g., Bolton and Lüning, 1982). Upper temperature limits on location we calculated the nearest distance (m) to the sea ice edge each growth of arctic kelps range from 16 to 21 °C (Assis et al., 2018), which week over the 10-year period. To compare these trends over this last are well above conditions found along Arctic coasts. This suggests decade with broader patterns of sea ice loss we obtained daily measures warming could more than double kelp production in some regions the of areal sea ice extent from NASA satellite data from 1980 to 2016 next 2–3 decades. Warming may also improve recruitment; for example, (Fig. 5). germination of spores, fertility (Golikov and Averintsev, 1977), and Of the total 1179 records of kelp, 2.6% occurred in locations where survival of arctic kelp gametophytes are limited by temperatures below the ice-free period was < 1 week in 2006 and 0.12% occurred where −1 °C (Sjøtun and Schoschina, 2002; Müller et al., 2008; Assis et al., the ice-free period was < 1 week in 2016 (mean 0.55 ± 0.99 SD), 2018). Such changes will vary across kelp species and will likely alter supporting evidence of survival and growth under extremely low light their competitive interactions. In the northern Gulf of Alaska, spore conditions (Wilce, 2016). On average, the annual mean and minimum settlement and gametophyte growth of E. fistulosa were more negatively distance (km) to sea ice (mean ± SD) were highly variable at kelp impacted by elevated temperatures and low salinity, than that of the locations (mean 221 ± 156 km and minimum 30 ± 62 km in 2006, more widely distributed N. luetkeana and S. latissima (Lind and Konar, and mean 274 ± 341 km and minimum 49 ± 138 km in 2016; Fig. 2017). A. esculenta is best adapted to low temperatures and cannot S2). For records that were under sea ice for at least 1 week during this survive in waters warmer than 16 °C (Sundene, 1962). Likewise, re- period, the mean distance to the sea ice edge increased from cruitment of L. solidungula becomes limited when temperatures exceed 45 ± 24 km to 88 ± 72 km and the minimum distance to sea ice edge 10 °C. Other, more warm adapted temperate kelps such as L. hyperborea, increased from 0.53 ± 1.52 km to 0.59 ± 1.88 km from 2006 to 2016. L. digitata and Saccharina polyschides may extend their range northward, Increases in distance to sea ice were largest in the White Sea and No- following the trend of boreal species moving into the Arctic (Fossheim vaya Zemlya, Russia and southeastern Greenland, and lowest in et al., 2015; Hargrave et al., 2017; Stige and Kvile, 2017). However, northern Canada and northeastern Russia (Fig. 5B). kelps produce short-lived zoospores that disperse slowly (current pat- Available evidence indicates that the loss of sea ice currently oc- terns of kelp diversity and structure can still be related to glacial cycles curring in the Arctic will lead to the northward expansion of kelps (Neiva et al., 2018), so any temperature-driven northern expansion of (Müller et al., 2009), and an increase in the depth range and pro- temperate kelp species into polar regions is likely to be slow (Konar, ductivity of these habitats due to increased light and reduced scour in 2007; Wilce, 2016). the surf zone, which narrows the vertical distribution of kelp (Krause- Jensen et al., 2012; Krause-Jensen and Duarte, 2014). Kelps cannot 5.2. Sea ice and light exist in areas with permanent sea ice (Shoshina et al., 2016), so ice loss may open new habitats in the high Arctic. The effect of sea ice loss on The amount of light reaching the benthos is a defining factor for kelps may even be stronger than anticipated because day length in- benthic primary production and depends largely on the extent of sea ice creases rapidly during the period of ice break-up (Clark et al., 2013), cover. Sea ice is rapidly retreating in the Arctic (areal loss of 3.5–4.5% implying a slight reduction in ice cover will result in a dis- per decade, Fig. 5A). Average sea ice extent ( ± SD) declined by 3.7% proportionately large increase in the amount of light reaching kelp. between 2006 and 2016 (from 16.2 ± 104 to 15.6 ± 105 M km2), and These expectations are supported by correlative studies from along the by 23% in 2016 compared to average sea ice measures from 1981 to west coast of Greenland showing that the extent of sea ice cover ex- 1989 (21.4 ± 2.4 M km2). plained 92% of the variation in maximum depth distribution and 80% To examine ongoing changes in sea ice extent at locations with re- of the variation in kelp growth (Krause-Jensen et al., 2012). Hop et al. cords of kelp, we obtained the position of the ice edge (defined by a (2012) monitored the biomass and depth range of kelps in Svalbard, threshold of > 15% sea ice cover) from NASA satellite images taken Norway between 1996 and 2014 and found that kelp biomass (mainly

Fig. 5. (A) Daily sea ice extent in millions of km for entire Arctic region between 1981 and 2010. (B) Change in mean distance to sea ice edge (km) between 2006 and 2016, for locations of kelp that occurred under ice for at least 1 week over this period.

9 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

L. digitata) recently increased 2–4 fold in the shallow zone (2.5 m to low transparency of water (< 3 m visibility) caused by domestic depth). They ascribed these changes to reductions in sea ice cover pollution, sediment plumes and agricultural run-off (Мalavenda and (Bartsch et al., 2016). Malavenda, 2012). These negative impacts may offset the possible po- sitive effects of warming and increased light on kelp growth insome Arctic regions. This was evident in the Beaufort Sea, where long-term 5.3. Salinity and turbidity records of annual growth of L. solidungula showed no change in pro- ductivity since 1979, despite earlier sea ice break-up and a longer ice- As a consequence of reduced sea ice and melting permafrost, many free period in recent years (Bonsell and Dunton, 2018). This pattern was Arctic coastlines are breaking apart and eroding into the sea. These explained by increasing resuspension of sediment and larger coastal traditionally icebound coasts can be fragile because ice provides pro- erosion following sea ice break-up, which counter balanced the positive tection from storms and waves, and its loss can expose the ground to the effect of longer ice-free periods. elements and make it unstable (Lantuit et al., 2012). Coastal environ- ments near these eroding regions are receiving higher amounts of se- diment loading and freshwater inputs, resulting in longer and more 5.4. Nutrients extreme periods of low salinity and intense turbidity and sedimentation (Lantuit et al., 2012; McClelland et al., 2012; Fritz et al., 2017). Since Nutrient concentrations are predicted to increase and change their 2000, average erosion rate of permafrost-bound coasts was 0.5 m yr−1, seasonal timing along Arctic coasts with increased (and earlier) spring and reached 10 m per yr−1 along some segments. Inputs of sediment melts, but the impacts of elevated nutrient richness on arctic kelps are and particulate organic carbon (POC) from coastal erosion are currently unclear. Nutrient availability is typically low in most Arctic waters, and entering the Arctic ocean at rates ~430 Tg yr−1 sediment and 4.9–14 nutrient concentrations tend to increase during winter when primary Tg yr−1 POC (Fritz et al., 2017). Coastal erosion is most severe along production is low, but decrease to extremely low levels during the short the shallow coasts of the Laptev, East Siberian and Beaufort Seas Arctic summer. Therefore, pelagic primary production is often limited (Lantuit et al., 2012), but increased turbidity from melting ice can also by low nutrient availability in late summer. be pronounced near the heads of Arctic fjords (Bartsch et al., 2016) and This may not be the case for kelps. In a study of twenty-one different in areas receiving glacial discharge (Traiger and Konar, 2018). species of arctic macroalgae (including Laminaria spp.), none of them Increased turbidity and reduced salinity is expected to reduce the were significantly nitrogen-limited in July (Gordillo et al., 2006). Kelps performance and lower depth limit of kelp by reducing light penetra- may be able to acquire and accumulate nutrients in winter when nu- tion and restricting photosynthesis (Aumack et al., 2007; Fredersdorf trient availability is relatively high. Nutrients can be translocated from et al., 2009; Spurkland and Iken, 2011; Wiencke and Amsler, 2012; the blade towards the meristem (Davison and Stewart, 1983) and nu- Traiger and Konar, 2018)(Fig. 6). Variable salinity reduced photo- trient reserves can subsequently be used to support photosynthesis and, synthetic efficiency of L. solidungula, S. dermatodea, L. digitata, A. es- thus, prolong blade growth during summer when insolation is high and culenta and S. latissima (Karsten, 2007). Laboratory experiments on nutrient availability is low (Gagne et al., 1982; Henley and Dunton, kelps collected from Svalbard, Norway found that sediment from 1997; Pueschel and Korb, 2001). Most kelp species should therefore melting ice negatively impacted their recruitment (Zacher et al., 2016). remain rather unaffected by increasing nutrient availability, but studies Manipulative field experiments on kelp forests in Alaska showed that have shown that the growth of at least some species, here L. solidungula, glacier run-off reduced kelp settlement and recruitment by increasing decreased significantly in early spring as nutrient concentrations sedimentation in the coastal zone (Traiger and Konar, 2018). Research dropped (Chapman and Lindley, 1980; Dunton et al., 1982). This sug- from Kola bay and anecdotal reports from areas along the Siberian shelf gests that some kelp species and/or kelps in extremely nutrient poor in Russia describe declines in the lower depth limit of kelp forests due areas can be limited by low nutrient availability, and would be

Fig. 6. Effects of environmental changes on arctic kelps from laboratory and field experiments. + is positive, − negative, 0 is no measurable effect, and? is unknown. Relative importance of stressors for 3 different coastal regions (see Fig. 2): ** = strong impact, * = moderate impact, and ‘x’ little to no impact. Note increased turbidity and decreased sali- nity can also occur along coasts with no sea ice that receive glacial melt or other freshwater inputs.

10 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14 stimulated by increased nutrient levels. region prevented them from concluding that no change had occurred. It is important to note that pelagic phytoplankton are more stimu- Northward range expansions of kelps may be limited by extensive gaps lated by increasing nutrient and light levels compared to benthic algae. between suitable substrate (e.g., from northern Norway to Svalbard) Estimates predict thus that the pelagic production by phytoplankton in and low dispersal potential of kelps (Wernberg et al., 2019). It is also some Arctic waters will increase 3-fold within this century due to longer possible that the spread and performance of kelps may be more influ- ice-free periods and increased run-off from land (e.g., Rysgaard and enced by changes in turbidity, sea ice cover, and light penetration Glud, 2007). This significant increase in phytoplankton biomass and compared to relatively small changes in sea temperatures. This suggests productivity will decrease light penetration in the water column, which that model predictions may overestimate northern range expansions of will negatively affect kelp biomass and depth limit, possibly offsetting kelps, at least in the short-term. any benefits that higher nutrient levels could have on some kelp spe- cies. 7. Conclusions

5.5. UV radiation The Arctic is at the epicenter of the global climate crisis, and emerging opportunities and developments have increased international Other changes in environmental conditions that could impact kelps attention on changes to ecosystems in this area. Long-term research include increased UV radiation, which is especially pronounced at high from Greenland and Norway suggests a warmer Arctic with less sea ice latitudes (Garcia-Corral et al., 2014). Increases in UV radiation nega- may support higher kelp productivity and biomass and expand the tively impacts photosynthesis of arctic kelps (Roleda et al., 2006; northern range and lower depth limit of these species. However, the Müller et al., 2008; Roleda, 2016) and reduces their performance degree to which these changes will positively affect kelps will vary (Heinrich et al., 2015). However, research to date indicates that UV regionally and depend on the extent that melting sea ice and permafrost damage will have a minor impact on arctic kelps compared other en- increases turbidity in coastal areas, as well as the available substrate in vironmental changes, and will mainly affect early life stages (Roleda the lower depth range (Bartsch et al. 2016; Bonsell and Dunton, 2018). et al., 2006; Wiencke et al., 2006). In laboratory experiments on L. Predictive models and laboratory experiments suggest the ‘borealiza- solidungula collected from Svaldbard by Roleda (2016), high UV ra- tion’ of arctic kelp forests will occur as temperatures warm, altering the diation disrupted the life cycle of meiospores and gametophytes. UV species composition of existing cold and ice-adapted kelp communities exposure also caused significant declines in photosynthetic efficiency, in high Arctic regions. Although current predictions are highly un- and increased transcription of DNA repair genes, but these effects were certain, the possible expansion of kelp forests should provide new ha- less pronounced in kelps collected from the field compared to cultured bitats for fish and other marine organisms, and a suite of valuable plants (Heinrich et al., 2015). Fredersdorf et al. (2009) examined ecosystem services along Arctic coastlines. Interestingly, where data are combined effects of different temperatures, salinity, and UV radiation available, kelp abundance appears relatively stable, suggesting these levels on photosynthesis of A. esculenta collected from Svalbard. They changes are occurring slower than predicted or are being buffered by found that A. esculenta zoospores were sensitive to synergistic effects of other factors. Either way, anticipating these changes, and under- temperature and salinity changes (Fredersdorf et al., 2009), but that standing these new ecosystems will be a key priority for northern adults could tolerate a range of UV conditions. communities. Our understanding of kelp forests is rapidly expanding in many 6. Predicting changes to distribution of arctic kelps regions of the Arctic. However, baseline measures of the extent of kelp communities are missing in northern and eastern Canadian Arctic, Predicting changes to arctic kelps under rapidly changing environ- Siberia, the east Greenland Shelf, and Russia. This lack of data is not mental conditions remains challenging. Assis et al. (2018) developed unique to kelp ecosystems. Despite the fact that over 28% of the world's models that described the current distributions of A. esculenta, L. soli- coastlines are found in the Arctic (Lantuit et al., 2012), they remain dungula, L. digitata, L. hyperborea, S. latissima, and S. dermatoada in the largely unstudied, which jeopardizes current strategies to protect or northern Atlantic according to environmental parameters (mainly sea conserve arctic environments and will have consequences for northern temperature, sea ice, salinity, upwelling), and used these relationships communities that rely on them. Lack of data has already greatly hin- to predict the impacts of climate change on their future distribution. dered our ability to detect and understand the impacts of climate These models predicted large northward expansions of these species, change on these and other ecosystems (e.g., Merzouk and Johnson, including the expansion of L. hyperborea to Svalbard, Norway, and 2011). Exploring effects of ongoing and future climate changes will further into the White Sea, the spread of S. dermatoada and L. digitata provide important insight on the stability of these ecosystems. Main- (or S. nigripes depending on source, S. Fredriksen personal commu- taining and augmenting current monitoring initiatives and time series nication) along the northeastern coast of Greenland, and the expansion data sets should be a priority. For kelp forests, understanding how these of A. esculenta into the Canadian high Arctic. The models also predicted ecosystems influence the structure and function of coastal arctic food L. solidungula and S. latissima would extend northward to cover the webs is an important focus for ongoing research. There is also a critical northernmost coasts of Greenland, Russia and Canada, suggesting that lack of knowledge on the contribution of kelp forests to carbon cycling all Arctic coasts would have environmental conditions suitable for kelp in the Arctic. Filling in these gaps and strategically prioritizing research forests in the future. Similar range expansions have been predicted for in areas of rapid environmental variation will enable us to more ef- L. solidungula and S. latissima with models by Müller et al. (2009) and fectively understand and conserve these ecosystems. for a number of fucoid species by Jueterbock et al. (2013, 2016). Arctic coasts are in line to become one of the most impacted en- However, there is a discrepancy between these predictive models and vironments in the world under changing climate. For this region to act long-term field observations of changes to arctic kelps. In Canada, Adey as a sentinel for climate change it is critical to monitor and understand and Hayek (2011) were unable to identify significant shifts in the dis- the impacts of environmental stressors on arctic ecosystems. Kelp for- tributions of subtidal algal species in the eastern subarctic or boreal ests provide a key example of the regional diversity of responses to regions over the past 40 years. Likewise, Merzouk and Johnson (2011) climate change, and demonstrate the need for a mechanistic under- reviewed the distribution of kelp along the northwest Atlantic shores standing of how multiple stressors and diverse ecological processes from records dating back to the 1950s and were unable to document influence ecosystem structure and function. Although it is tempting to any significant change in dominant kelp species composition or abun- make generalized statements about broad-scale climate-driven impacts, dance over that period, despite increasing sea temperature, although, the reality is much more nuanced, regionally specific, and highly un- the lack of sufficient spatially and temporally extensive datasets for this certain. What is clear is that extensive ecological changes are likely to

11 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14 occur in these rapidly changing environments, with both ‘positive’ or Clark, G.F., Stark, J.S., Johnston, E.L., et al., 2013. Light-driven tipping points in polar ‘negative’ consequences for a range of species. ecosystems. Glob. Chang. Biol. 19, 3749–3761. https://doi.org/10.1111/gcb.12337. Cooper, L.W., Beasley, T.M., Zhao, X.-L., et al., 1998. Iodine-129 and plutonium isotopes in Arctic kelp as historical indicators of transport of nuclear fuel-reprocessing wastes Acknowledgements from mid-to-high latitudes in the Atlantic Ocean. Mar. Biol. 131, 391–399. https:// doi.org/10.1007/s002270050332. Coupel, P., Ruiz-Pino, D., Sicre, M.A., et al., 2015. The impact of freshening on phyto- This work was funded by the Norwegian Research Council through plankton production in the Pacific Arctic Ocean. Prog. Oceanogr. 131, 113–125. the KELPEX project (NRC grant no. 255085). In addition, TW received https://doi.org/10.1016/J.POCEAN.2014.12.003. funding from The Australian Research Council (DP170100023). We are Davison, I.R., Stewart, W.D.P., 1983. Occurrence and significance of nitrogen transport in grateful for comments from Eva Ramirez Llodra and two anonymous the brown alga Laminaria digitata. Mar. Biol. 77, 107–112. https://doi.org/10.1007/ BF00396307. reviewers. Dayton, P.K., 1975. Experimental studies of algal canopy interactions in a sea otter- dominated kelp community at Amchitka Island,Alaska. Fish Bull. 73, 230–237. Appendix A. Supplementary data Dayton, P.K., 2013. Polar benthos. In: Smith, W.O. (Ed.), Polar Oceanography: Chemistry, Biology, and Geology. Academic Press, pp. 614. Dean, T.A., Bodkin, J.L., Jewett, S.C., et al., 2000a. Changes in sea urchins and kelp Supplementary data to this article can be found online at https:// following a reduction in sea otter density as a result of the Exxon Valdez oil spill. Mar. doi.org/10.1016/j.gloplacha.2018.09.005. Ecol. Prog. Ser. 199, 281–291. Dean, T.A., Haldorson, L., Laur, D.R., et al., 2000b. The distribution of nearshore fishes in kelp and eelgrass communities in Prince William Sound, Alaska: associations with References vegetation and physical habitat characteristics. Environ. Biol. Fish 57, 271–287. https://doi.org/10.1023/A:1007652730085. Ding, Q., Schweiger, A., L'Heureux, M., et al., 2017. Influence of high-latitude atmo- Abdullah, M.I., Fredriksen, S., Christie, H., 2017. The impact of the kelp (Laminaria hy- spheric circulation changes on summertime Arctic sea ice. Nat. Clim. Chang. 7, perborea) forest on the organic matter content in sediment of the west coast of 289–295. https://doi.org/10.1038/nclimate3241. Norway. Mar. Biol. Res. 13, 151–160. https://doi.org/10.1080/17451000.2016. Doroff, A.M., Estes, J.A., Tinker, M.T., et al., 2003. Sea otter population declines inthe 1240369. Aleutian archipelago. J. Mammal. 84, 55–64. https://doi.org/10.1644/1545- Acosta Navarro, J.C., Varma, V., Riipinen, I., et al., 2016. Amplification of Arctic warming 1542(2003)084<0055:SOPDIT>2.0.CO;2. by past air pollution reductions in Europe. Nat. Geosci. 9, 277–281. https://doi.org/ Duggins, D.O., 1980. Kelp beds and sea otters: an experimental approach. Ecology 61, 10.1038/ngeo2673. 447–453. Adey, W.H., Hayek, L.-A.C., 2011. Elucidating marine biogeography with macrophytes: Duggins, D.O., Simenstad, C.A., Estes, J.A., 1989. Magnification of secondary production quantitative analysis of the North Atlantic supports the thermogeographic model and by kelp detritus in coastal marine ecosystems. Science 80 (245), 170–173. demonstrates a distinct subarctic Region in the Northwestern Atlantic. Northeast. Dunton, K.H., 1985. Growth of dark-exposed Laminaria saccharina (L.) Lamour. and Nat. 18, 1–128. https://doi.org/10.1656/045.018.m801. Laminaria solidungula J. Ag. (laminariales : phaeophyta) in the Alaskan Beaufort Sea. AMAP, 1998. Physical/geographical characteristics of the Arctic. In: AMAP Assessment J. Exp. Mar. Biol. Ecol. 94, 181–189. https://doi.org/10.1016/0022-0981(85) Report: Arctic Pollution Issues. Arctic Monitoring and Assessment Programme 90057-7. (AMAP), Oslo, pp. 9–24. Dunton, K.H., Jodwalis, C.M., 1988. Photosynthetic performance of Laminaria solidungula AMAP, 2017. Snow, Water, Ice and Permafrost in the Arctic (SWIPA) 2017. Arctic measured in situ in the Alaskan High Arctic. Mar. Biol. 98, 277–285. https://doi.org/ Monitoring and Assessment Programme (AMAP), Oslo, Norway (Oslo). 10.1007/BF00391206. Arctic Monitoring and Assessment Programme, 2011. Snow, Water, Ice and Permafrost in Dunton, K.H., Schell, D.M., 1987. Dependence of consumers on macroalgal (Laminaria the Arctic (SWIPA): Climate Change and the Cryosphere. AMAP. solidungula) carbon in an arctic kelp community: 13C evidence. Mar. Biol. 93, Assis, J., Araújo, M.B., Serrão, E.A., 2018. Projected climate changes threaten ancient 615–625. https://doi.org/10.1007/BF00392799. refugia of kelp forests in the North Atlantic. Glob. Chang. Biol. 24, e55–e66. https:// Dunton, K.H., Schell, D.M., 1986. Seasonal carbon budget and growth of Laminaria soli- doi.org/10.1111/gcb.13818. dungula in the Alaskan High Arctic. Mar. Ecol. Prog. Ser. 31, 57–66. https://doi.org/ Aumack, C.F., Dunton, K.H., Burd, A.B., et al., 2007. Linking light attenuation and sus- 10.3354/meps031057. pended sediment loading to benthic productivity within an arctic kelp-bed commu- Dunton, K.H., Reimnitz, E., Schonberg, S., 1982. An arctic kelp community in the Alaskan nity. J. Phycol. 43, 853–863. https://doi.org/10.1111/j.1529-8817.2007.00383.x. Beaufort Sea. Arctic 35, 465–484. Bartsch, I., Paar, M., Fredriksen, S., et al., 2016. Changes in kelp forest biomass and depth Estes, J.A., Duggins, D.O., 1995. Sea otters and kelp forests in Alaska: generality and distribution in Kongsfjorden, Svalbard, between 1996–1998 and 2012–2014 reflect variation in a community ecological paradigm. Ecol. Monogr. 65, 75–100. https:// Arctic warming. Polar Biol. 39, 2021–2036. https://doi.org/10.1007/s00300-015- doi.org/10.2307/2937159. 1870-1. Estes, J.E., Smith, N.S., Palmisano, J.F., 1978. Sea otter predation and community or- von Biela, V.R., Newsome, S.D., Bodkin, J.L., et al., 2016. Widespread kelp-derived ganization in the Western Aleutian Islands, Alaska. Ecology 59, 822–833. https://doi. carbon in pelagic and benthic nearshore fishes suggested by stable isotope analysis. org/10.2307/1938786. Estuar. Coast. Shelf Sci. 181, 364–374. https://doi.org/10.1016/J.ECSS.2016.08. Estes, J.A., Danner, E.M., Doak, D.F., et al., 2004. Complex trophic interactions in kelp 039. forest ecosystems. Bull. Mar. Sci. 74, 621–638. Boertmann, D., Mosbech, A., Schiedek, D., Dünweber, M., 2013. Disko West: A Strategic Fagerli, C., Norderhaug, K., Christie, H., 2013. Lack of sea urchin settlement may explain Environmental Impact Assessment of Hydrocarbon Activities. Aarhus University, DCE kelp forest recovery in overgrazed areas in Norway. Mar. Ecol. Prog. Ser. 488, – Danish Centre for Environment and Energy. 119–132. https://doi.org/10.3354/meps10413. Bolton, J.J., Lüning, K., 1982. Optimal growth and maximal survival temperatures of Fagerli, C.W., Stadniczeñko, S.G., Pedersen, M.F., et al., 2015. Population dynamics of Atlantic Laminaria species (Phaeophyta) in culture. Mar. Biol. 66, 89–94. https://doi. Strongylocentrotus droebachiensis in kelp forests and barren grounds in Norway. Mar. org/10.1007/BF00397259. Biol. 162, 1215–1226. https://doi.org/10.1007/s00227-015-2663-3. Bonsell, C., Dunton, K.H., 2018. Long-term patterns of benthic irradiance and kelp pro- Filbee-Dexter, K., Scheibling, R.E., 2014. Sea urchin barrens as alternative stable states of duction in the central Beaufort Sea reveal implications of warming for Arctic inner collapsed kelp ecosystems. Mar. Ecol. Prog. Ser. 495, 1–25. https://doi.org/10.3354/ shelves. Prog. Oceanogr. https://doi.org/10.1016/j.pocean.2018.02.016. meps10573. Borum, K., Pedersen, M.F., Krause-Jensen, D., Christensen, N., 2002. Biomass, photo- Filbee-Dexter, K., Wernberg, T., 2018. Rise of Turfs: a new battlefront for globally de- synthesis and growth of Laminaria saccharina in a high-arctic fjord, NE Greenland. clining kelp forests. Bioscience 68, 64–76. https://doi.org/10.1093/biosci/bix147. Mar. Biol. 141, 11–19. https://doi.org/10.1007/s00227-002-0806-9. Filbee-Dexter, K., Feehan, C.J., Scheibling, R.E., 2016. Large-scale degradation of a kelp Bringloe, T., Dunton, K.H., Saunders, G.W., 2017. Updates to the marine algal flora of the ecosystem in an ocean warming hotspot. Mar. Ecol. Prog. Ser. 543, 141–152. https:// Boulder Patch in the Beaufort Sea off northern Alaska as revealed by DNA barcoding. doi.org/10.3354/meps11554. Arctic 70, 343–348. https://doi.org/10.14430/arctic4679. Filbee-Dexter, K., Wernberg, T., Ramirez-Llodra, E., et al., 2018. Movement of pulsed Burrows, M.T., Schoeman, D.S., Buckley, Lauren B., et al., 2011. The pace of shifting resource subsidies from shallow kelp forests to deep fjords. Oecologia 187, 291–304. climate in marine and terrestrial ecosystems. Science 334, 652–655. https://doi.org/ Floeter, S.R., Behrens, M.D., Ferreira, C.E.L., et al., 2005. Geographical gradients of 10.1126/science.1210288. marine herbivorous fishes: patterns and processes. Mar. Biol. 147, 1435–1447. Catton, C., 2016. “Perfect Storm” Decimates Northern California Kelp Forests | CDFW https://doi.org/10.1007/s00227-005-0027-0. Marine Management News. Calif. Dep. Fish Wildl. Mar. Manag. News. https:// Fossheim, M., Primicerio, R., Johannesen, E., et al., 2015. Recent warming leads to a cdfwmarine.wordpress.com/2016/03/30/perfect-storm-decimates-kelp/, Accessed rapid borealization of fish communities in the Arctic. Nat. Clim. Chang. 5, 673–677. date: 25 February 2018. https://doi.org/10.1038/nclimate2647. Chapman, A.R.O., Lindley, J.E., 1980. Seasonal growth of Laminaria solidungula in the Fredersdorf, J., Müller, R., Becker, S., et al., 2009. Interactive effects of radiation, tem- Canadian High Arctic in relation to irradiance and dissolved nutrient concentrations. perature and salinity on different life history stages of the Arctic kelp Alaria esculenta Mar. Biol. 57, 1–5. https://doi.org/10.1007/BF00420961. (Phaeophyceae). Oecologia 160, 483–492. https://doi.org/10.1007/s00442-009- Christie, H., Gundersen, H., Rinde, E., Bekkby, T., 2014. In: Laminaria Hyperborea Kelp 1326-9. Forest as an Indicator in "Nature index of Norway". Norwegian Institute for Water Fredriksen, S., 2003. Food web studies in a Norwegian kelp forest based on stable isotope Research (NIVA-rapport, 6609), pp. 34 ISBN 978-82-577-6344-2. analysis. Mar. Ecol. Prog. Ser. 260, 71–81. https://doi.org/10.3354/meps260071. Christie, H., Norderhaug, K.M., Fredriksen, S., 2009. Macrophytes as habitat for fauna. Fritz, M., Vonk, J.E., Lantuit, H., 2017. Collapsing Arctic coastlines. Nat. Clim. Chang. 7, Mar. Ecol. Prog. Ser. 396, 221–234. 6–7. https://doi.org/10.1038/nclimate3188.

12 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

Gagne, J.A., Mann, K.H., Chapman, A.R.O., 1982. Seasonal patterns of growth and storage Konar, B., Edwards, M.S., Bland, A., et al., 2017. A swath across the great divide: kelp in Laminaria longicruris in relation to differing patterns of availability of nitrogen in forests across the Samalga Pass biogeographic break. Cont. Shelf Res. 143, 78–88. the water. Mar. Biol. 69, 91–101. https://doi.org/10.1007/BF00396965. https://doi.org/10.1016/j.csr.2017.06.007. Garcia-Corral, L.S., Agustí, S., Regaudie-de-Gioux, A., et al., 2014. Ultraviolet radiation Kortsch, S., Primicerio, R., Fossheim, M., et al., 2015. Climate change alters the structure enhances arctic net plankton community production. Geophys. Res. Lett. 41, of arctic marine food webs due to poleward shifts of boreal generalists. Proc. Biol. Sci. 5960–5967. https://doi.org/10.1002/2014GL060553. 282, 20151546. https://doi.org/10.1098/rspb.2015.1546. Gattuso, J.-P., Gentili, B., Duarte, C.M., et al., 2006. Light availability in the coastal Krause-Jensen, D., Duarte, C.M., 2014. Expansion of vegetated coastal ecosystems in the ocean: impact on the distribution of benthic photosynthetic organisms and con- future Arctic. Front. Mar. Sci. 1. https://doi.org/10.3389/fmars.2014.00077. tribution to primary production. Biogeosciences 3, 489–513. https://doi.org/10. Krause-Jensen, D., Duarte, C.M., 2016. Substantial role of macroalgae in marine carbon 5194/bg-3-489-2006. sequestration. Nat. Geosci. 9, 737–742. https://doi.org/10.1038/ngeo2790. Golikov, A.N., Averintsev, V., 1977. Biotsenozy verhnih otdelov shelfa arhipelaga Zemlya Krause-Jensen, D., Kühl, M., Christensen, P., Borum, J., 2007. Benthic primary production Frantsa Iosifa [Biocenosis of the high part of archipelago Franz Josef land shelf]. in Young Sound, Northeast Greenland. In: Ryssguard, S., Glud, R.N. (Eds.), Benthic Issled Fauny Morey 14, 5–54. Primary Production in Young Sound, Northeast Greenland. Bioscience, Meddr, Gordillo, F.J.L., Aguilera, J., Jiménez, C., 2006. The response of nutrient assimilation and Grønland, pp. 160–173. biochemical composition of Arctic seaweeds to a nutrient input in summer. J. Exp. Krause-Jensen, D., Marbà, N., Olesen, B., et al., 2012. Seasonal sea ice cover as principal Bot. 57, 2661–2671. https://doi.org/10.1093/jxb/erl029. driver of spatial and temporal variation in depth extension and annual production of Gotceitas, V., Fraser, S., Brown, J.A., 1995. Habitat use by juvenile Atlantic cod (Gadus kelp in Greenland. Glob. Chang. Biol. 18, 2981–2994. https://doi.org/10.1111/j. morhua) in the presence of an actively foraging and non-foraging predator. Mar. Biol. 1365-2486.2012.02765.x. 123, 421–430. https://doi.org/10.1007/BF00349220. Krumhansl, K., Scheibling, R., 2012. Production and fate of kelp detritus. Mar. Ecol. Prog. Grebmeier, J.M., Overland, J.E., Moore, S.E., et al., 2006. A major ecosystem shift in the Ser. 467, 281–302. https://doi.org/10.3354/meps09940. northern Bering Sea. Science 311, 1461–1464. https://doi.org/10.1126/science. Krumhansl, K.A., Okamoto, D.K., Rassweiler, A., et al., 2016. Global patterns of kelp 1121365. forest change over the past half-century. Proc. Natl. Acad. Sci. 113, 13785–13790. Guiry, M., Guiry, G., 2017. Database. Natl. Univ. Ireland Galw. https://doi.org/10.1073/pnas.1606102113. Gulliksen, B., Sandnes, O., 1980. Marine bunndyrsamfunn, nøkkelarter og felteksper- Küpper, F.C., Peters, A.F., Shewring, D.M., et al., 2016. Arctic marine phytobenthos of imenter på hardbunn. Fauna 33, 19. northern Baffin Island. J. Phycol. 52, 532–549. https://doi.org/10.1111/jpy.12417. Gulliksen, B., Haug, T., Sandnes, O., 1980. Benthic macrofauna on new and old lava Kuznetsov, L., Shoshina, E., 2003. Barents Sea Phytocenosises (Physiological and grounds at Jan Mayen. Sarsia 65, 13745. Structural Characteristics). 308 p. Apatity Publ house KSC RAS. Günther, F., Overduin, P.P., Yakshina, I.A., et al., 2015. Observing Muostakh disappear: Kuznetsov, L., Makarevich, P., Makarov, M., 1994. Structural-condition-indicators of permafrost thaw subsidence and erosion of a ground-ice-rich island in response to marine phytocenoses // habitat and ecosystems of Franz Josef Land (Archipelago and arctic summer warming and sea ice reduction. Cryosphere 9, 151–178. https://doi. shelf). Apatity 89–94. org/10.5194/tc-9-151-2015. Lantuit, H., Overduin, P.P., Couture, N., et al., 2012. The Arctic coastal dynamics data- Halpern, B.S., Walbridge, S., Selkoe, K.A., et al., 2008. A global map of human impact on base: a new classification scheme and statistics on arctic permafrost coastlines. marine ecosystems. Science 319, 948–952. https://doi.org/10.1126/science. Estuar. Coasts 35, 383–400. https://doi.org/10.1007/s12237-010-9362-6. 1149345. Lee, R.K.S., 1973. General ecology of the Canadian Arctic benthic marine algae. Arctic 26, Hamilton, J., Brenda, K., 2007. Implications of substrate complexity and kelp variability 32–43. for south-central Alaskan nearshore fish communities. Fish. Bull. 105, 189–196. Lee, R.K.S., 1980. A Catalogue of the Marine Algae of the Canadian Arctic. National Hanelt, D., 1998. Capability of dynamic photoinhibition in Arctic macroalgae is related to Museums of Canada. their depth distribution. Mar. Biol. 131, 361–369. https://doi.org/10.1007/ van Leeuwe, M.A., Tedesco, L., Arrigo, K.R., et al., 2018. Microalgal community structure s002270050329. and primary production in Arctic and Antarctic sea ice: a synthesis. Elem. Sci. Anth. Hargrave, M.S., Foggo, A., Pessarrodona, A., Smale, D.A., 2017. The effects of warming on 6. https://doi.org/10.1525/elementa.267. the ecophysiology of two co-existing kelp species with contrasting distributions. Leinaas, H.P., Christie, H., 1996. Effects of removing sea urchins (Strongylocentrotus Oecologia 183, 531–543. https://doi.org/10.1007/s00442-016-3776-1. droebachiensis): stability of the barren state and succession of kelp forest recovery in Harley, C.D.G., Anderson, K.M., Demes, K.W., et al., 2012. Effect of climate change on the East Atlantic. Oecologia 105, 524–536. https://doi.org/10.1007/BF00330016. global seaweed communities. J. Phycol. 48, 1064–1078. https://doi.org/10.1111/j. Leont'yev, I.O., 2003. Modeling erosion of sedimentary coasts in the western Russian 1529-8817.2012.01224.x. Arctic. Coast. Eng. 47, 413–429. https://doi.org/10.1016/S0378-3839(02)00145-X. Heinrich, S., Valentin, K., Frickenhaus, S., Wiencke, C., 2015. Temperature and light Levin, V., Muller, O., Anisimova, N., 1998. Part II Bottom invertebrates. In: Harvesting interactively modulate gene expression in Saccharina latissima (Phaeophyceae). J. and Prospective Algae and Invertebrates for Uses of the Barents and White Seas. Kola Phycol. 51, 93–108. https://doi.org/10.1111/jpy.12255. Science Centre Apatity, Murmansk, pp. 394–440. Henley, W.J., Dunton, K.H., 1997. Effects of nitrogen supply and continuous darkness on Lind, A.C., Konar, B., 2017. Effects of abiotic stressors on kelp early life-history stages. growth and photosynthesis of the arctic kelp Laminaria solidungula. Limnol. Algae 32, 223–233. https://doi.org/10.4490/algae.2017.32.8.7. Oceanogr. 42, 209–216. https://doi.org/10.4319/lo.1997.42.2.0209. Ling, S.D., Johnson, C.R., Ridgeway, K., et al., 2009. Climate-driven range extension of a Hobday, A.J., Pecl, G.T., 2014. Identification of global marine hotspots: sentinels for sea urchin: inferring future trends by analysis of recent population dynamics. Glob. change and vanguards for adaptation action. Rev. Fish Biol. Fish. 24, 415–425. Chang. Biol. 15, 719–731. https://doi.org/10.1111/j.1365-2486.2008.01734.x. https://doi.org/10.1007/s11160-013-9326-6. Makarov, M., Ryizhik, I., Voskoboynikov, G., 2008. Mehanizmy suschestvovaniya buryh Hop, H., Wiencke, C., Vögele, B., Kovaltchouk, N.A., 2012. Species composition, zonation, vodorosley v period polyarnoy nochi: funktsionalnaya differentsiatsiya i geterotrofiya and biomass of marine benthic macroalgae in Kongsfjorden, Svalbard. Bot. Mar. 55, [Mechanisms of existence of brown seaweeds during polar night: functional differ- 399–414. https://doi.org/10.1515/bot-2012-0097. entiation and heterotrophy]. Sovrem. Probl. Algol. 225–227. Huang, J., Zhang, X., Zhang, Q., et al., 2017. Recently amplified arctic warming has Мalavenda, C., Malavenda, C., 2012. Черты деградации в фитоценозах южного и contributed to a continual global warming trend. Nat. Clim. Chang. 7, 875–879. среднего колен Кольского залива Баренцева моря. Bull. Moscow State Tech. Univ. https://doi.org/10.1038/s41558-017-0009-5. 15, 794–802. IPCC, 2014. In: Core Writing Team, Pachauri, R.K., Meyer, L.A. (Eds.), Climate Change McClelland, J.W., Holmes, R.M., Dunton, K.H., Macdonald, R.W., 2012. The Arctic ocean 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth estuary. Estuar. Coasts 35, 353–368. https://doi.org/10.1007/s12237-010-9357-3. Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva. McDevit, D.C., Saunders, G.W., 2010. A DNA barcode examination of the Laminariaceae Jueterbock, A., Tyberghein, L., Verbruggen, H., et al., 2013. Climate change impact on (Phaeophyceae) in Canada reveals novel biogeographical and evolutionary insights. seaweed meadow distribution in the North Atlantic rocky intertidal. Ecol. Evol. 3, Phycologia 49, 235–248. https://doi.org/10.2216/PH09-36.1. 1356–1373. https://doi.org/10.1002/ece3.541. McMeans, B., Rooney, N., Arts, M., Fisk, A., 2013. Food web structure of a coastal Arctic Jueterbock, A., Smolina, I., Coyer, J.A., Hoarau, G., 2016. The fate of the arctic seaweed marine ecosystem and implications for stability. Mar. Ecol. Prog. Ser. 482, 17–28. Fucus distichus under climate change: an ecological niche modeling approach. Ecol. https://doi.org/10.3354/meps10278. Evol. 6, 1712–1724. https://doi.org/10.1002/ece3.2001. Merzouk, A., Johnson, L.E., 2011. Kelp distribution in the Northwest Atlantic Ocean Karsten, U., 2007. Research note: salinity tolerance of arctic kelps from Spitsbergen. under a changing climate. J. Exp. Mar. Biol. Ecol. 400, 90–98. https://doi.org/10. Phycol. Res. 55, 257–262. https://doi.org/10.1111/j.1440-1835.2007.00468.x. 1016/j.jembe.2011.02.020. Kawai, H., Hanyuda, T., Ridgway, L.M., Holser, K., 2013. Ancestral reproductive structure Miller, A.W., Ruiz, G.M., 2014. Arctic shipping and marine invaders. Nat. Clim. Chang. 4, in basal kelp Aureophycus aleuticus. Sci. Rep. 3, 2491. https://doi.org/10.1038/ 413–416. https://doi.org/10.1038/nclimate2244. srep02491. Müller, R., Wiencke, C., Bischof, K., 2008. Interactive effects of UV radiation and tem- Keats, D.W., South, G.R., Steele, D.H., 1985. Algal biomass and diversity in the upper perature on microstages of Laminariales (Phaeophyceae) from the Arctic and North subtidal at a pack-ice disturbed site in eastern Newfoundland. Mar. Ecol. Prog. Ser. Sea. Clim. Res. 37, 203–213. 25, 151–158. Müller, R., Laepple, T., Bartsch, I., Wiencke, C., 2009. Impact of oceanic warming on the Kjellman, F., 1883. The Algae of the Arctic Sea: A Survey of the Species, Together with an distribution of seaweeds in polar and cold-temperate waters. Bot. Mar. 52. https:// Exposition of the General Characters and the Development of the Flora. (Norstedt). doi.org/10.1515/BOT.2009.080. Konar, B., 2007. Recolonization of a high latitude hard-bottom nearshore community. Mundy, C.J., Ehn, J.K., Barber, D.G., Michel, C., 2007. Influence of snow cover and algae Polar Biol. 30, 663–667. https://doi.org/10.1007/s00300-007-0261-7. on the spectral dependence of transmitted irradiance through Arctic landfast first- Konar, B., 2013. Lack of recovery from disturbance in high-arctic boulder communities. year sea ice. J. Geophys. Res. 112, C03007. https://doi.org/10.1029/2006JC003683. Polar Biol. 36, 1205–1214. https://doi.org/10.1007/s00300-013-1340-6. Myagkov, G.M., 1975. Composition, distribution and the seasonal dynamics of algal Konar, B., Estes, J.A., 2003. The stability of boundary regions between kelp forests and biomass in the Laminaria biocenosis of the bay of the White Sea USSR based on diving deforested areas. Ecology 84, 174–185. https://doi.org/10.1890/0012-9658(2003) data. Bull. Leningr. State Univ. Ser. Biol. 3, 48–53. 084[0174:TSOBRB]2.0.CO;2. Najafi, M.R., Zwiers, F.W., Gillett, N.P., 2015. Attribution of Arctic temperature changeto

13 K. Filbee-Dexter et al. Global and Planetary Change 172 (2019) 1–14

greenhouse-gas and aerosol influences. Nat. Clim. Chang. 5, 246–249. https://doi. SRES, 2000. IPCC Special Report: Summary for Policymakers Emissions Scenarios. org/10.1038/nclimate2524. Stige, L.C., Kvile, K.Ø., 2017. Climate warming drives large-scale changes in ecosystem Neiva, J., Paulino, C., Nielsen, M.M., et al., 2018. Glacial vicariance drives phylogeo- function. Proc. Natl. Acad. Sci. U. S. A. 114, 12100–12102. https://doi.org/10.1073/ graphic diversification in the amphi-boreal kelp Saccharina latissima. Sci. Rep. 8, pnas.1717090114. 1112. https://doi.org/10.1038/s41598-018-19620-7. Sundene, o, 1962. The implications of transplant and culture experiments on the growth Nelson, R.J., Ashjian, C.J., Bluhm, B.A., et al., 2014. Biodiversity and biogeography of the and distribution of Alaria esculenta. Nytt. Mag. Bot. 9, 155–174. lower trophic taxa of the pacific Arctic region: sensitivities to climate change. In:The Svendsen, P., 1959. The algal vegetation of Spitsbergen: a survey of the marine algal flora Pacific Arctic Region. Springer Netherlands, Dordrecht, pp. 269–336. of the outer part of isfjorden. Nor. Polarinst. Skr. 116, 1–49. Nordenskiøld, A., 1880. Vegas fård kring Asien och Europa jemte en historisk å terblick på Taylor, W.R., 1954. Algae: non-planktonic. Bot. Rev. 20, 363–399. føregående resor længs Gamla Verldens nordkust. F & G Beijers Førlag, Stockholm. Teagle, H.A., Hawkins, S.J., Moore, P., Smale, D.A., 2017. The role of kelp species as Norderhaug, K.M., Christie, H., 2011. Secondary production in a Laminaria hyperborea biogenic habitat formers in coastal marine ecosystems. J. Exp. Mar. Biol. Ecol. 492, kelp forest and variation according to wave exposure. Estuar. Coast. Shelf Sci. 95, 81–98. https://doi.org/10.1016/J.JEMBE.2017.01.017. 135–144. https://doi.org/10.1016/J.ECSS.2011.08.028. Traiger, S.B., Konar, B., 2018. Mature and developing kelp bed community composition in Parkinson, C.L., Comiso, J.C., 2013. On the 2012 record low Arctic sea ice cover: com- a glacial estuary. J. Exp. Mar. Biol. Ecol. 501, 26–35. https://doi.org/10.1016/J. bined impact of preconditioning and an August storm. Geophys. Res. Lett. 40, JEMBE.2017.12.016. 1356–1361. https://doi.org/10.1002/grl.50349. Wang, K., Zhang, T., Zhang, X., et al., 2017. Continuously amplified warming in the Pecl, G.T., Hobday, A.J., Frusher, S., et al., 2014. Ocean warming hotspots provide early Alaskan Arctic: implications for estimating global warming hiatus. Geophys. Res. warning laboratories for climate change impacts. Rev. Fish Biol. Fish. 24, 409–413. Lett. 44, 9029–9038. https://doi.org/10.1002/2017GL074232. https://doi.org/10.1007/s11160-014-9355-9. Wassmann, P., Reigstad, M., 2011. Future arctic ocean seasonal ice zones and implica- Phillips, R.L., Reiss, T.E., 1985. Nearshore Marine Geologic Investigations, Point Barrow tions for pelagic-benthic coupling. Oceanography 24, 220–231. https://doi.org/10. to Skull Cliff, Northeast Chukchi Sea. (Menlo Park). 5670/oceanog.2011.74. Piepenburg, D., 2005. Recent research on Arctic benthos: common notions need to be Waters, C.N., Zalasiewicz, J., Summerhayes, C., et al., 2016. The Anthropocene is func- revised. Polar Biol. 28, 733–755. https://doi.org/10.1007/s00300-005-0013-5. tionally and stratigraphically distinct from the Holocene. Science 351, 137–148. Plotkin, A.S., Railkin, A.I., Gerasimova, E.I., et al., 2005. Subtidal underwater rock https://doi.org/10.1126/science.aad2622. communities of the White Sea: structure and interaction with bottom flow. Russ. J. Wein, E.E., Freeman, M.M.R., Makus, J.C., 1996. Use of and preference for traditional Mar. Biol. 31, 335–343. https://doi.org/10.1007/s11179-006-0001-9. foods among the Belcher Island Inuit. Arctic 49, 256–264. Pueschel, C.M., Korb, R.E., 2001. Storage of nitrogen in the form of protein bodies in the Wernberg, T., Bennett, S., Babcock, R.C., et al., 2016. Climate-driven regime shift of a kelp Laminaria solidungula. Mar. Ecol. Prog. Ser. 218, 107–114. https://doi.org/10. temperate marine ecosystem. Science 353, 169–172. https://doi.org/10.1126/ 3354/meps218107. science.aad8745. Reedy, K., Katherine, 2016. Kelp-fed beef, swimming caribou, feral reindeer, and their Wernberg, T., Krumhansl, K.A., Filbee-Dexter, K., Pedersen, M.F., 2019. Status and trends hunters: island mammals in a marine economy. Sustainability 8, 113. for the world’s kelp forests. In: Sheppard, C. (Ed.), World Seas: An Environmental Renaud, P.E., Løkken, T.S., Jørgensen, L.L., et al., 2015. Macroalgal detritus and food-web Evaluation. Vol. III Elsevier. subsidies along an Arctic fjord depth-gradient. Front. Mar. Sci. 2, 31. https://doi.org/ Wiencke, C., Amsler, C.D., 2012. Seaweeds and Their Communities in Polar Regions. 10.3389/fmars.2015.00031. Springer, Berlin, Heidelberg, pp. 265–291. Roleda, M.Y., 2016. Stress physiology and reproductive phenology of Arctic endemic kelp Wiencke, C., Clayton, M.N., 2009. Biology of polar benthic algae. Bot. Mar. 52. https:// Laminaria solidungula J. Agardh. Polar Biol. 39, 1967–1977. https://doi.org/10.1007/ doi.org/10.1515/BOT.2009.083. s00300-015-1813-x. Wiencke, C., Clayton, M.N., 2011. Introduction: biology of polar benthic algae. In: Roleda, M.Y., Hanelt, D., Wiencke, C., 2006. Exposure to ultraviolet radiation delays Wiencke, C. (Ed.), Biology of Polar Benthic Algae. De Gruyter, Berlin, pp. 337. photosynthetic recovery in Arctic kelp zoospores. Photosynth. Res. 88, 311–322. Wiencke, C., Roleda Michael, Y., Gruber, A., et al., 2006. Susceptibility of zoospores to UV https://doi.org/10.1007/s11120-006-9055-y. radiation determines upper depth distribution limit of Arctic kelps: evidence through Rosenvinge, L.K., 1893. Grønland havalger. Medd. Grønland 3, 763–981. field experiments. J. Ecol. 94, 455–463. https://doi.org/10.1111/j.1365-2745.2006. Rosenvinge, L.K., 1899. Deuxième mémoire sur les algues marines du groenland. Reitzel, 01102.x. København. Wilce, R., 1963. Studies on benthic marine algae in north-west Greenland. In: Proc Int Rysgaard, S., Glud, R.N., 2007. Carbon cycling and climate change: predictions for a High Seaweed Symp, pp. 280–287. Arctic marine ecosystem (Young Sound, NE Greenland). In: Rysgaard, S., Glud, R.N. Wilce, R.T., 2016. The “Arctic Stamp”, its imprint on an endangered marine flora. (Eds.), Carbon cycling in Arctic marine ecosystems: Case study Young Sound. Perspect. Phycol. 3, 155–180. https://doi.org/10.1127/pip/2016/0046. Bioscience, Meddr, pp. 206–214. Wilce, R.T., Dunton, K.H., 2014. The Boulder Patch (North Alaska, Beaufort Sea) and its Sharp, G., Allard, M., Lewis, A., et al., 2008. The potential for seaweed resource devel- Benthic Algal Flora. Arct. Inst. North Am. 67, 43–56. opment in subarctic Canada; Nunavik, Ungava Bay. J. Appl. Phycol. 20, 491–498. Wilmers, C.C., Estes, J.A., Edwards, M., et al., 2012. Do trophic cascades affect the storage https://doi.org/10.1007/s10811-008-9323-7. and flux of atmospheric carbon? An analysis of sea otters and kelp forests. Front.Ecol. Shoshina, E., Anisimova, N., 2013. Makrovodorosli iz rayona buhty Ledyanaya Gavan Environ. 10, 409–415. https://doi.org/10.1890/110176. (Novaya Zemlya, o. Severnyi, Karskoe more) [Seaweeds from Ledjanaya Haven Inlet Włodarska-Kowalczuk, M., Kukliński, P., Ronowicz, M., et al., 2009. Assessing species (Novaya Zemlja, Northern Island, Kara Sea)]. Vestn. MGTU 16, 530–535. richness of macrofauna associated with macroalgae in Arctic kelp forests (hornsund, Shoshina, E., Makarov, V., Makarov, M., 1997. Biological features of laminarians of svalbard). Polar Biol. 32, 897–905. https://doi.org/10.1007/s00300-009-0590-9. Frantz Josef Land. Biol. Morya 5, 286–292. Wulff, A., Iken, K., Quartino, M.L., et al., 2011. Biodiversity, biogeography and zonation Shoshina, E.V., Kapkov, V.I., Belenikina, O.A., 2016. Ecological factors regulating growth of marine benthic micro- and macroalgae in the Arctic and Antarctic. In: Wiencke, C. of seaweeds in Arctic communities. Вестник МГТУ 19, 334–344. (Ed.), Biology of Polar Benthic Algae. De Gruyter, pp. 337. Sjøtun, K., Schoschina, E.V., 2002. Gametophytic development of Laminaria spp. Zacher, K., Bernard, M., Bartsch, I., Wiencke, C., 2016. Survival of early life history stages (Laminariales, Phaeophyta) at low temperature. Phycologia 41, 147–152. https:// of Arctic kelps (Kongsfjorden, Svalbard) under multifactorial global change scenarios. doi.org/10.2216/i0031-8884-41-2-147.1. Polar Biol. 39, 2009–2020. https://doi.org/10.1007/s00300-016-1906-1. Spurkland, T., Iken, K., 2011. Kelp bed dynamics in estuarine environments in subarctic Zenkevitch, L., 1963. Biology of the Seas of the U.S.S.R. J. Chem. Inf. Model. 1–955. Alaska. J. Coast. Res. 275, 133–143. https://doi.org/10.2112/JCOASTRES-D-10- https://doi.org/10.1017/CBO9781107415324.004. 00194.1.

14